PRELIMINARY REPORT ON PEAT.

HERBS

Panicum hemitomon (maiden cane) Eriocaulon compressum Andropogon sp. (broom-sedge) Lycopodium Chapmani Pontederia cordata (wampee) Drosera capillaris Spartina Bakeri (a large grass) Anchistea Virginica (a fern) Eupatorium capillifolium (dog-fennel)
Castalia odorata (white water-lily)
(stunted)
Centella repanda Osmunda cinnamomea (at fern) Rhynchospora axillaris Fuirena scirpoidea Solidago fistulosa (golden-rod)

Amphicarpum sP. (a grass) Sagittaria lancifolia Erianthus sp. (a tall grass) Cladium effusum (saw-grass) Eleocharis interstincta Rhynchospora Tracyi Triadenum Virginicum Bartonia verna Gyrotheca tinctoria (paint-root) Rhynchosporo. fascicularis Syngonanthus flavidulus Tillandsia usneoides (Spanish moss) Tillandsia recurvata (air-plant) Gerardia linifolia Osmunda regalis (a fern) Lorinseria areolata (a fern) Dryopteris Thelypteris (a fern)

MOSSES

Sphagnum sp.

The trees, vines and air-plants and nearly all the shrubs and ferns grow in dense clumps which are a very characteristic feature of the larger peat prairies. (See illustration.) The location of these clumps seems to have little to do with the depth of the peat under them, and in fact I have not been able to discover any striking correlation between the depth of various parts of these prairies (whose solid sandy bottoms are sometimes very irregular) and the vegetation on the surface, except that the saw-grass and a few other plants seem to be confined to the wettest places. It will be noticed that nearly all the plants mentioned a few pages back as characteristic of sandy shores of small lakes occur also in these peat prairies.
The peat of the prairies of course contains a few woody roots and little or no sediment, and it is just about the purest peat we have. (See analyses under localities 4, 13, 15, 29 and 41, and miscellane-ous No. i). One sample from Lake County (No. 29.11) showed only i 2 %o of ash, which is the purest peat of which we have any record. Depths of 6 or 8 feet in these prairies are common, and some exceed 20 feet (which is considerably more than is needed to work with a dredge to the greatest advantage). Where samples have been taken from different depths in the same prairie those from near the surface usually show more mineral matter than those

gffs